Neurodegenerative disorders are a leading cause of morbidity and mortality, and the burden of such disorders is expected to grow dramatically over the coming decades. Despite this, there is currently no effective therapeutics for prevention or treatment of most neurodegenerative conditions. Nucleotide repeat expansions are a common cause of neurodegeneration and neurological illness, affecting millions of people worldwide. Recently, nucleotide repeats have been found to promote a novel form of translational initiation known as Repeat-Associated Non-AUG (RAN) translation, producing homo-polymeric or dipeptide repeat containing proteins by initiating translation in the absence of an AUG start codon. Our group has recently discovered that RAN translation occurs in association with CGG repeat expansions within the FMR1 gene that cause the neurodegenerative disorder Fragile X-associated tremor/ataxia syndrome (FXTAS). Our preliminary data demonstrates that CGG RAN translation produces both poly-glycine and poly-alanine containing proteins from separate open reading frames. Production of the poly-glycine RAN product contributes to toxicity in simple model systems and accumulates in neuronal inclusions in FXTAS patient brain tissue, supporting a central role for this process in FXTAS pathogenesis. However, how exactly CGG repeats elicit RAN translation and what factors of the general translation machinery are required for these processes is currently unknown. Based on our preliminary data, we hypothesize that CGG RAN translation can occur by at least two separate novel mechanisms that are dependent on the utilized reading frame. First, we propose that expanded CGG repeats can stall scanning pre-initiation complexes (PICs) in the 5'UTR of FMR1, allowing for initiation at a near-AUG start codon (such as GUG) to produce the poly-glycine protein. In contrast, we propose that CGG repeat expansions can also directly recruit translational initiation machinery and trigger initiatin within the repeat itself to produce the poly-alanine protein. To test these translational initiatio models, we will assess the requirements for the following three key steps of canonical translation initiation through in vitro and cell culture-based experiments: 1) 5' cap recognition, ) mRNA unwinding and PIC scanning, and 3) start codon selection. Aim 1 will determine if CGG RAN translation in each reading frame is cap-dependent. Aim 2 will determine if mRNA unwinding and PIC scanning by RNA helicases and repeat-destabilizing proteins influence CGG RAN translation. Aim 3 will determine the role of initiation factors in non-AUG start codon selection. We will then extend these studies to determine the impact of modulating these same pathways has on CGG repeat elicited toxicity in Drosophila. These studies will identify the mechanisms by which RAN translation allows for translational initiation in different frames at expanded CGG repeats, while providing critical insights into pathogenesis and therapeutic development for FXTAS and other neurodegenerative repeat expansion disorders.